Q1: Let X 1 , … , X n X_1,\dots,X_n X1,…,Xn be an iid sample of N ( 1 , 2 2 ) N(1,2^2) N(1,22). Which of the following items are true? ( D )
A. X ˉ − 1 2 / n ∼ t ( n ) \frac{\bar X-1}{2/\sqrt{n}}\sim t(n) 2/nXˉ−1∼t(n)
B. 1 4 ∑ i = 1 n ( X i − 1 ) 2 ∼ F ( n , 1 ) \frac{1}{4}\sum_{i=1}^n(X_i-1)^2\sim F(n,1) 41∑i=1n(Xi−1)2∼F(n,1)
C. X ˉ − 1 2 / n ∼ N ( 0 , 1 ) \frac{\bar X-1}{\sqrt{2}/\sqrt{n}}\sim N(0,1) 2/nXˉ−1∼N(0,1)
D. 1 4 ∑ i = 1 n ( X i − 1 ) 2 ∼ χ 2 ( n ) \frac{1}{4}\sum_{i=1}^n(X_i-1)^2\sim \chi^2(n) 41∑i=1n(Xi−1)2∼χ2(n)
Q2: Let X 1 , … , X n X_1,\dots,X_n X1,…,Xn be a simple random sample of normal population N ( μ , σ 2 ) N(\mu,\sigma^2) N(μ,σ2).
-
Find the mean and variance of S n 2 S_n^2 Sn2.
-
Show that S n 2 ∼ G a m m a ( ( n − 1 ) / 2 , n / ( 2 σ 2 ) ) S_n^2\sim Gamma((n-1)/2,n/(2\sigma^2)) Sn2∼Gamma((n−1)/2,n/(2σ2)).
Solution: 已知 Y : = n S n 2 / σ 2 ∼ χ 2 ( n − 1 ) Y:=nS_n^2/\sigma^2\sim\chi^2(n-1) Y:=nSn2/σ2∼χ2(n−1),因此 E [ Y ] = n − 1 , V a r [ Y ] = 2 ( n − 1 ) \mathbb E[Y]=n-1,\quad Var[Y]=2(n-1) E[Y]=n−1,Var[Y]=2(n−1)令 S n 2 = σ 2 n Y S_n^2=\frac{\sigma^2}{n}Y Sn2=nσ2Y则有 E [ S n 2 ] = σ 2 n E [ Y ] = ( n − 1 ) σ 2 n V a r [ S n 2 ] = σ 4 n 2 V a r [ Y ] = 2 ( n − 1 ) σ 4 n 2 \mathbb E[S_n^2]=\frac{\sigma^2}{n}\mathbb E[Y]=(n-1)\frac{\sigma^2}{n}\\ Var[S_n^2]=\frac{\sigma^4}{n^2}Var[Y]=2(n-1)\frac{\sigma^4}{n^2} E[Sn2]=nσ2E[Y]=(n−1)nσ2Var[Sn2<